Anthropic has restored access to its most powerful AI models after an eighteen-day suspension triggered by US export control regulations. The company now offers Claude Sonnet 5 alongside restored versions of its Fable and Mythos frontier models, following a federal review that concluded on June 30.
The pause began June 12 when US government officials issued an export control directive targeting Anthropic's highest-capability systems. The restriction prevented international users from accessing these models, forcing Anthropic to temporarily disable them while navigating compliance requirements.
Export controls on advanced AI reflect broader US government efforts to restrict access to frontier AI capabilities abroad, particularly to countries of concern. These regulations target systems that officials believe could pose national security risks if deployed outside US oversight. The eighteen-day window suggests federal reviewers determined Anthropic's approach to deployment and access controls met export compliance standards.
Claude Sonnet 5 represents Anthropic's latest evolution in its model lineup. The company positions Sonnet as a mid-tier offering between its smaller Claude Haiku models and more demanding Opus variants. Fable and Mythos, now restored, sit at the frontier capability tier and serve enterprise customers requiring maximum performance.
The restoration matters because export restrictions create friction in the AI market. International companies, researchers, and developers relying on Anthropic's models faced service disruptions during the pause. Resuming operations signals that Anthropic's governance frameworks satisfied federal requirements without requiring architectural changes to how the company operates.
The incident underscores how AI companies navigate an evolving regulatory landscape. Unlike traditional tech exports, AI models raise unique compliance challenges. Regulators must balance national security interests against the need for continued innovation and international collaboration. Anthropic's quick resolution suggests the company's voluntary compliance measures and access controls aligned with government expectations.
This development also indicates US officials view Anthropic as a responsible actor capable of